INDEX

Explanations

expressions of positive or negative emotional reactions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

aver

-0.16

ordes

-0.15

-0.14

/from

-0.14

acades

-0.14

iry

-0.14

NR

-0.14

adge

-0.13

edException

-0.13

afort

-0.13

POSITIVE LOGITS

ness

0.20

NESS

0.19

HeaderCode

0.16

GuidId

0.16

.nih

0.16

 KrÃ¡l

0.15

 ENTRY

0.14

-looking

0.14

iye

0.14

 capt

0.14

Activations Density 0.156%

expressions of positive or negative emotional reactions

No Comments

No Known Activations